Hydrolysis of acetylthiocoline, o-nitroacetanilide and o-nitrotrifluoroacetanilide by fetal bovine serum acetylcholinesterase Marı ´ a F. Montenegro, Marı ´ a T. Moral-Naranjo, Encarnacio ´ n Mun ˜ oz-Delgado, Francisco J. Campoy and Cecilio J. Vidal Departamento de Bioquı ´ mica y Biologı ´ a Molecular-A, Universidad de Murcia, Spain Although the best defined function of cholinesterases (ChEs) is the cleavage of acetylcholine (ACh) in syn- aptic and nonsynaptic locations, both acetylcholines- terase (AChE; EC 3.1.1.7) and butyrylcholinesterase (BuChE; EC 3.1.1.8) possess the capacity to hydrolyze o-nitroacetanilides by means of the so-called aryl acylamidase (AAA) activity [1–3]. Despite the lack of information on the natural substrate and physiologi- cal significance of the amidase activity of ChEs, its variation during zebrafish embryogenesis [4] and brain development [5] and its possible involvement in Alzhei- mer’s disease [6] make the amidase activity a matter of interest. AChE and BuChE have 53% sequence homology, similar folding of subunits, and similar quaternary structures. Both ChEs show a wide range of molecular forms, which arise from transcriptional, post-transcrip- tional and post-translational changes [7,8]. The AChE Keywords aryl acylamidase; chemical denaturation; cholinesterases; kinetic parameters; molecular forms Correspondence C. J. Vidal, Departamento de Bioquı ´ mica y Biologı ´ a Molecular-A, Edificio de Veterinaria, Universidad de Murcia, Apdo. 4021, E-30071 Espinardo, Murcia, Spain Fax: +34 968 364147 Tel: +34 968 364774 E-mail: cevidal@um.es (Received 12 November 2008, revised 27 January 2009, accepted 30 January 2009) doi:10.1111/j.1742-4658.2009.06942.x Besides esterase activity, acetylcholinesterase (AChE) and butyrylcholinest- erase (BuChE) hydrolyze o-nitroacetanilides through aryl acylamidase activity. We have reported that BuChE tetramers and monomers of human blood plasma differ in o-nitroacetanilide (ONA) hydrolysis. The homology in quaternary structure and folding of subunits in the prevalent BuChE species (G H 4 ) of human plasma and AChE forms of fetal bovine serum prompted us to study the esterase and amidase activities of fetal bovine serum AChE. The k cat ⁄ K m values for acetylthiocholine (ATCh), ONA and its trifluoro derivative N-(2-nitrophenyl)-trifluoroacetamide (F-ONA) were 398 · 10 6 m )1 Æmin )1 , 0.8 · 10 6 m )1 Æmin )1 , and 17.5 · 10 6 m )1 Æmin )1 , respectively. The lack of inhibition of amidase activity at high F-ONA con- centrations makes it unlikely that there is a role for the peripheral anionic site (PAS) in F-ONA degradation, but the inhibition of ATCh, ONA and F-ONA hydrolysis by the PAS ligand fasciculin-2 points to the transit of o-nitroacetalinides near the PAS on their way to the active site. Sedimenta- tion analysis confirmed substrate hydrolysis by tetrameric 10.9S AChE. As compared with esterase activity, amidase activity was less sensitive to guan- idine hydrochloride. This reagent led to the formation of 9.3S tetramers with partially unfolded subunits. Their capacity to hydrolyze ATCh and F-ONA revealed that, despite the conformational change, the active site architecture and functionality of AChE were partially retained. Abbreviations AAA, aryl acylamidase activity; ACh, acetylcholine; AChE, acetylcholinesterase; ATCh, acetylthiocholine iodide; Brij 96, polyoxyethylene 10 - oleyl ether; BuChE, butyrylcholinesterase; BuTCh, butyrylthiocholine; BW (BW284c51), 1,5-bis(4-allyldimethylammonium phenyl)-pentan-3- one dibromide; ChE, cholinesterase; Fas2, fasciculin-2; F-ONA, N-(2-nitrophenyl)-trifluoroacetamide; GPI, glycosyl phosphatidylinositol; Iso- OMPA, tetraisopropyl pyrophosphoramide; Nbs 2, 5,5¢-dithiobis(2-nitrobenzoic acid); ONA, o-nitroacetanilide; PAS, peripheral anionic site. 2074 FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS gene produces several types of mature mRNA, depend- ing on the choice of splice acceptor site in the 3¢-region of the primary transcript. Thus, the AChE-T (‘tailed’), AChE-H (‘hydrophobic’) and AChE-R (‘read- through’) mRNAs generate the three principal types of AChE subunit, referred to as T, H, and R, which dif- fer in their C-termini. Also, AChE subunits can have different N-terminal regions [9]. Upon polymerization, AChE subunits generate various molecular forms, which can be classified as globular (G) and asymmetric (A) forms. The globular forms may exhibit hydrophilic (G H ) or amphiphilic (G A ) properties, according to the folding of the C-terminal domain in the monomers (G 1 ) and dimers (G 2 ) composed of AChE-T subunits, the absence or presence of the hydrophobic proline- rich membrane anchor (PRiMA subunit) in the G 4 species composed of AChE-T subunits, and the addi- tion of glycosyl phosphatidylinositol (GPI) to the G 1 and G 2 species consisting of AChE-H subunits [7]. Crystallographic techniques and site-directed muta- genesis [10] have revealed a deep narrow gorge in AChE that penetrates halfway into the protein and contains the catalytic site at 4 A ˚ from its base. About 70% of the gorge is lined by 14 aromatic residues, which are in the various subsites involved in ACh accommodation. These include Trp86, Tyr133, Tyr337 and Phe338 (numbers for human AChE) of the ‘anionic subsite’, which contributes to the stabilization of the quaternary ammonium function of the choline moiety, and Trp236, Phe295 and Phe297 of the ‘acyl-binding subsite’, which fits the acetyl group into a concave hydrophobic pocket. The structural elements involved in the fitting of ACh in the Michaelis complex, stabil- ization of the tetrahedral intermediate and hydrolysis are Trp86 of the anionic binding site, the ‘oxyanion hole’ (Gly121, Gly129, Ala204), which stabilizes the negative charge created at the carbonyl oxygen atom of ACh in the acetylation–deacetylation process, and the catalytic triad (Ser203, Glu334, His447), where the Ser hydroxyl group acts as the nucleophile against the car- bonyl group of ACh, and the Glu-His pair is crucial for activation of the nucleophile. At the rim of the gorge, Tyr72, Tyr124, Trp286 and Tyr341 form part of the ‘peripheral anionic site’ (PAS), which can bind ACh as well as fasciculin-2 (Fas2) and the second qua- ternary ammonium of decamethonium. Crystallography data [11] have shown that BuChE has a catalytic triad (Ser198, Glu325, His438) near the bottom of a 20 A ˚ gorge, which is lined by only eight aromatic amino acid residues, an acyl-binding pocket (including Trp231, Leu286, Phe329, and Val288), an anionic subsite (Trp82 and Phe329), an oxyanion hole (Gly116, Gly117, and Ala199), and a nonaromatic guide (Asp70 and Tyr332), which pushes butyryl- thiocholine (BuTCh) down the active site gorge. We have previously reported that BuChEs of human colon, kidney and serum exhibit varying amidase ⁄ esterase activity ratios [12]. The same applies for BuChEs of chick, horse, and fetal bovine serum [13]. In liver pathologies, there is an important increase in amidase activity of human serum BuChE [14]. The difference between BuChE monomers and tetra- mers in their capacity to hydrolyze o-nitroacetanilide (ONA) and its trifluoro derivative N-(2-nitrophenyl)- trifluoroacetamide (F-ONA) led us to attribute the varying amidase ⁄ esterase activity ratios to the molec- ular polymorphism of BuChE and subtle structural differences in the subunits [12]. AChE and BuChE subunits generate a similar range of oligomers by poly- merization. The variable extents to which ONA and F-ONA are hydrolyzed by G 1 and G 4 BuChE species from human sources [12] prompted us to investigate the esterase and amidase activities of purified fetal bovine serum AChE. For this purpose, catalytic effi- ciencies of AChE acting on esterase [acetylthiocholine (ATCh)] and amidase substrates (ONA and F-ONA) were determined. In addition, hydrolysis of the sub- strates by different molecular forms of AChE was compared. Finally, the spatial requirements of AChE subunits for expressing esterase and amidase activities were probed by their response to the protein denatur- ant guanidine hydrochloride. Results and Discussion Kinetic parameters of amidase activity associated with AChE from fetal bovine serum Affinity-purified AChE from fetal bovine serum expressed both amidase and esterase activities. Both were inhibited to a similar extent by increasing concen- trations of 1,5-bis(4-allyldimethylammonium phenyl)- pentan-3-one dibromide (BW284c51; BW) and Fas2 (Fig. 1). This feature, and the binding of esterase and amidase activities to antibodies (HR2) against AChE (Fig. 1), ruled out any contribution of BuChE and possible contaminating esterases to the hydrolysis of F-ONA and ONA. Hence, the results shown hereafter correspond exclusively to esterase and amidase activi- ties of AChE. The almost complete suppression of both amidase and esterase activities by the PAS inhibitor Fas-2 and the active site inhibitor BW (Fig. 1) corroborates the involvement of the same active site in the hydrolysis of esterase (ATCh) and amidase (ONA and F-ONA) substrates, a fact that has been established since M. F. Montenegro et al. Aryl acylamidase activity of AChE FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 2075 Moore & Hess [15] demonstrated the equal effects of pH on K m and k cat values for esterase and amidase activities and their irreversible inhibition by an organo- phosphate. Affinity-purified fetal bovine serum AChE was able to hydrolyze 319 ± 12 nmol ONA per mg of protein per minute (about 30 nmolÆmin )1 ÆmL )1 ), provided that amidase activity was assayed at 10 mm ONA. The activity was 20-fold that reported for purified AChE from monkey brain [16]. In this regard, it is worth mentioning the proposal of the high level of AChE in early brain development as the source for the abundant amidase activity in mammalian fetal serum [5]. In addition, some authors have implicated amidase activity in the formation of senile plaques through embryonic- like AChE species identified in Alzheimer brain [6]. The catalytic parameters of AChE obtained by tracing the esterase (ATCh) and amidase (F-ONA and ONA) substrates are given in Table 1. The results show that fetal bovine serum AChE displays a lower K m value with ATCh than with F-ONA and ONA, a trait also observed for BuChE [17]. Although K m can- not be strictly considered to be a reliable measure of substrate affinity, the lower K m of AChE with ATCh than with o-nitroacetanilides probably reflects the obstacles encountered by amidase substrates in reach- ing and binding to the active site, which are due in part to their nitro group (see later). The nearly 19-fold lower K m value for fetal bovine serum AChE with F-ONA than with ONA (Table 1) suggests that the three fluorine atoms improve accessibility and binding of F-ONA to the acyl-binding site. The calculated K m value for fetal bovine serum AChE with ATCh as substrate (0.23 mm) is similar to those reported for AChE of bovine (0.22 mm) and human (0.28 mm) erythrocytes [18], and electric tissue Fig. 1. Esterase and amidase activities of purified fetal bovine serum AChE are due to AChE. (A) Effect of AChE inhibitors on esterase and amidase activities of fetal bovine serum AChE. Inhibition of esterase activity, assayed with 1 m M ATCh, and of amidase activity, measured with 10 m M ONA and 0.3 mM F-ONA, by BW and Fas2. (B) Binding of esterase and amidase activities to antibodies against AChE bound to protein G–agarose. Table 1. Kinetic parameters for esterase and amidase activities of AChE purified from fetal bovine serum. In the case of fetal bovine serum AChE, the content of catalytic sites in the reaction mixture was determined with echothiophate. In assays with ATCh, the content of cata- lytic sites was 70 ± 10 n M, and in assays with F-ONA and ONA it was 1540 nM. The much lower hydrolysis rate of nitroacetanilides than of ATCh makes it necessary to raise the amount of AChE in F-ONA and ONA assays for accurate determination of the extent of their hydroly- sis. Figures represent mean values of triplicate determinations. Constant values reported for cholinesterases from other sources are included for comparison. Enzyme Substrate K m (mM) k cat (min )1 ) k cat ⁄ K m (M )1 Æmin )1 ) Reference Fetal bovine serum AChE ATCh 0.23 ± 0.01 91 700 ± 3700 398 ± 34 · 10 6 This work Fetal bovine serum AChE F-ONA 0.81 ± 0.01 14 200 ± 100 17.5 ± 0.3 · 10 6 This work Fetal bovine serum AChE ONA 15.22 ± 2.24 12 500 ± 2400 0.8 ± 0.3 · 10 6 This work Human AChE ATCh 0.1 400 000 4.0 · 10 9 [20] Human BuChE BuTCh 0.018 24 000 1.3 · 10 9 [42] Human BuChE F-ONA > 3 > 7500 2.5 · 10 6 [26] Human BuChE ONA > 3 > 13 4.4 · 10 3 [26] Aryl acylamidase activity of AChE M. F. Montenegro et al. 2076 FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS of Electrophorus (0.23–0.32 mm) [19], but higher than that for recombinant human AChE (0.1 mm) [20]. The differences in K m of AChE according to animal species and tissues could arise from changes in active site structure, owing to amino acid sequence, quaternary structure (G 1 ,G 2 ,G 4 ), and post-translational changes, such as glycosylation and incorporation of GPI residues. These factors could also explain the species- specific differences in the responses of brain AChE to organophosphorus insecticides [21]. The difference between dimeric GPI-anchored AChEs of bovine erythrocytes and of lymphocytes in both K m and extent of lectin binding [22] illustrates the impact of oligoglycans on the kinetic behavior of AChE. The range of ATCh concentration, the ionic strength in the assay and the kinetic model used for fitting experimen- tal data may also contribute to the difference in K m . Regarding substrate inhibition, fetal bovine serum AChE was inhibited by ATCh at above 2 mm, but amidase activity was unaffected even at 5 mm F-ONA (more than 10 times the concentrations used in stan- dard assays; the poor solubility of F-ONA in water prevented us from using larger concentrations). The lack of inhibition of amidase activity at 5 mm F-ONA makes it unlikely that there is a role for the PAS in increasing the association rate constant. This conten- tion is supported by the absent or weak effects of the PAS inhibitors propidium, gallamine and decametho- nium on hydrolysis of ONA by electric eel AChE [23]. Nevertheless, the absence of a role for the PAS in o-nitroacetanilide binding does not necessarily imply an alternative route by which the catalytic pocket can be reached. In fact, the almost complete abolition of ATCh, ONA and F-ONA hydrolysis in the presence of Fas2 (Fig. 1), a PAS-binding inhibitor [24], demon- strates that o-nitroacetanilides transit near the PAS on their way to the active site. Thus, it seems that the ste- ric blockade created by a large polypeptide such as Fas2 at the PAS [25] would render AChE unable to hydrolyze both esterase and amidase substrates. The values of k cat in Table 1 indicate that F-ONA and ONA are degraded by fetal bovine serum AChE at almost the same rate, which is much lower than the rate for ATCh. The comparable activity of AChE on both amidase substrates contrasts with the much faster hydrolysis of F-ONA than ONA by BuChE [12,17,26]. The calculated k cat value for fetal bovine serum AChE with ATCh as the substrate (91 700 min )1 ) is much lower than that for AChE of humans (400 000 min )1 ) [20] and Electrophorus (350 500 min )1 ) [27]. These k cat values are all greater than for the hydrolysis of F-ONA (14 200 min )1 ) and ONA (12 500 min )1 ) (Table 1). The much lower k cat for AChE with aryl acetamides than with ATCh could arise from the greater energy required for breaking amide than ester bonds. Nevertheless, the greater k cat values for AChE (see above) than for BuChE on F-ONA and ONA (> 7500 min )1 and > 13 min )1 , respectively) [26] indi- cate that o-nitroacetanilides are hydrolyzed more rapidly by AChE. The moderate difference in k cat val- ues for AChE and BuChE with F-ONA, as compared with the large difference with ONA, underlines the low catalytic rate of ONA hydrolysis by BuChE, and the need for the three fluorine atoms in F-ONA and their activating effect on the carbonyl group for increasing the rate of turnover by BuChE. The faster hydrolysis of F-ONA than of ONA by BuChE has been attri- buted to differences in the acetylation rate [26]. As regards the orientation of o-nitroacetanilides in the active site of ChEs, it has been suggested that they adopt a planar and rigid structure, because of their aromatic ring, amide bond, partial electron delocaliza- tion, and hydrogen bonding between NH and NO ) [17,26]. In addition, molecular modeling suggests that the rigid structure of ONA orients its NO 2 group towards the oxyanion hole of BuChE [26]. If this is true, its occupation by the NO 2 group would presum- ably prevent binding and stabilization of the alkoxide ion of the tetrahedral transition state, and conse- quently would impair production of the acylated tran- sition state. Although it is not known whether ONA and F-ONA adopt the same orientation in the BuChE active site, the lower activation energy (higher k cat ⁄ K m ) with F-ONA than with ONA may reflect the contribu- tion of the fluorine atoms in F-ONA to the stabiliza- tion of the alkoxide moiety and further acetylation of BuChE. As the oxyanion hole of BuChE is composed of Gly116, Gly117, and Ala199, and the corresponding AChE hole is composed of Gly121, Gly122, and Ala204 [28], the faster hydrolysis of ONA by AChE would not arise from the amino acids in the hole, but possibly from improved accommodation and stabiliza- tion of the negative alkoxide intermediate in the AChE oxyanion site. It might be thought that the narrower active site in AChE than in BuChE would hamper ONA hydrolysis by AChE. Nevertheless, we feel that a smaller cavity could force the fitting and stabilization of ONA in the acyl-binding pocket, and by this means facilitate the production of the alkoxide group and its entry into the oxyanion hole. The narrower space in AChE may allow enzyme–substrate contacts that have a positive effect on catalysis. In contrast, a more loose interaction of ONA with the larger acyl-binding pocket of BuChE may prevent the substrate from assuming the required conformation for producing the alkoxide group. M. F. Montenegro et al. Aryl acylamidase activity of AChE FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 2077 When k cat ⁄ K m is used as a measure of catalytic effi- ciency, the results in Table 1 indicate that AChE works 23 times more efficiently with ATCh than with F-ONA, and about 500 times better than with ONA. The greater efficiency of AChE in acting on F-ONA than on ONA confirms the profitable use of F-ONA for assessing the amidase activity of AChE and BuChE [12]. AChE works approximately 22-fold more effi- ciently with F-ONA than with ONA (compare k cat ⁄ K m for the two substrates in Table 1) and BuChE 570-fold more efficiently [26]. These data agree with previous observations showing that the electron-withdrawing action of the three fluorine atoms on F-ONA, which are absent on ONA, lowers the energy barrier for breakage of the CO–NH bond by ChEs [17,26]. The lower activation energy (570-fold higher k cat ⁄ K m ) for F-ONA hydrolysis than for hydrolysis of ONA by BuChE arises from the big jump in the turnover of F-ONA (> 7500 min )1 ) as compared to that of ONA (> 13 min )1 ) [26]. To summarize, the hydrolysis of ATCh, F-ONA and ONA by AChE confirms its capacity for degrading esterase and amidase substrates. Nevertheless, the higher catalytic efficiency with ATCh emphasizes substrate preference and the main physiological role of AChE. Expression of esterase and amidase activities by AChE molecular forms We have previously reported that F-ONA and ONA are not hydrolyzed by BuChE tetramers of human gut, so that dimers and monomers are solely responsible for their hydrolysis [12]. In the case of plasma, whereas ONA was principally degraded by BuChE monomers and to a much lower extent by tetrameric species, both BuTCh and F-ONA were hydrolyzed by monomers and tetramers [12]. With these premises in mind, sedi- mentation analysis was undertaken to test the behavior of AChE forms in fetal bovine serum. Sedimentation profiles showed that the hydrophilic AChE tetramers (G H 4 ) are the principal species in fetal bovine serum (Fig. 2), in agreement with previous results [29]. Assays of amidase activity showed that, in contrast to the BuChE tetramers of human gut and blood plasma [12], the tetramers of fetal bovine serum had the capacity to hydrolyze ONA and F-ONA (Fig. 2). A shoulder at 8.6S (Fig. 2) and a small peak at 5.5S were occasionally observed in amidase activity profiles. They probably correspond to proteolytically trimmed tetramers and dimers, which are hardly detected in esterase activity profiles. If so, their obser- vation in assays with F-ONA and ONA suggested a higher amidase ⁄ esterase activity ratio for G 2 than for G 4 AChE, an idea in agreement with the results of Boopathy & Layer [5], who showed higher ONA- hydrolyzing capacity for G 1 and G 2 than for G 4 AChE from developing chicken brain. The capacity of G H 4 AChE of fetal bovine serum to hydrolyze F-ONA and ONA, the ability of G H 4 BuChE of human plasma to degrade F-ONA but not ONA, and the inability of G H 4 BuChE of human gut to degrade F-ONA and ONA, despite the structural homology between AChE and BuChE subunits, illus- trate how variable the hydrolysis of amidase substrates by ChE tetramers can be. The accessibility of the cata- lytic site to substrate would depend on the structure of the selected substrate, different amino acids in AChE and BuChE polypeptides, and subtle structural varia- tions between AChE (or BuChE) subunits according to the homotetramer source (animal species, tissue, and biological fluid). Post-translational changes in ChE subunits, e.g. those arising from oligoglycans, may affect peptide backbone fluctuations ⁄ flexibility, and by this means hamper the ability of ONA and ⁄ or F-ONA to enter the catalytic gorge, while the ability Fig. 2. Sedimentation patterns showing hydrolysis of esterase and amidase substrates by fetal bovine serum AChE. Sucrose density fractions were collected and assayed with ATCh, ONA, and F-ONA. Note the shoulder in the assays with amidase substrates and its absence in assays with ATCh. Aryl acylamidase activity of AChE M. F. Montenegro et al. 2078 FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS of ATCh to enter is maintained. In addition, active site occlusion may arise from steric hindrance by a nearby subunit in tetramers composed of tightly packed subunits [30]. We are currently investigating possible differences between human AChE monomers, dimers and tetramers in their capacity to hydrolyze esterase and amidase substrates. Effect of guanidine hydrochloride on esterase and amidase activities In agreement with our previous data [29], a dramatic reduction of ATCh-hydrolyzing activity was observed when fetal bovine serum AChE was exposed to 2 m guanidine hydrochloride (Fig. 3). As expected, the ami- dase activity was reduced to a similar extent. Nevertheless, the remaining activity sufficed for testing whether esterase and amidase activities differed in sensitivity to higher guanidine hydrochloride concen- trations. Above 2 m guanidine hydrochloride, the rate of ATCh hydrolysis fell rapidly and that of F-ONA degradation fell gradually, which provided a peak in the amidase ⁄ esterase activity ratio (Fig. 3). As both activities reside in the same protein, we believe that the unequal amidase ⁄ esterase activity ratio could arise from the conformational rearrangement at high guani- dine hydrochloride concentrations, so that the struc- tural change could generate molecules lacking esterase activity and retaining amidase activity or, alternatively, molecular forms with unequally impaired esterase and amidase activities. Fig. 4. Sedimentation profiles showing the effect of guanidine hydrochloride on the molecular distribution of esterase and amidase activities of fetal bovine serum AChE. Samples incubated without and with 2.3 M guanidine hydrochloride were centrifuged on 5–20% sucrose gradi- ents containing 0.5 M guanidine hydrochloride and 0.5% Brij 96, formed on top of concentrated sucrose (40%) to prevent possible AChE aggregates from reaching the tube bottom. Esterase activity was assayed with 1 m M ATCh (left graph) and amidase activity with 0.3 mM F-ONA (right). Note the different scales used for control and guanidine hydrochloride-treated fetal bovine serum AChE (left axis and right axis, respectively). The results reveal the formation of 9.3S species at the expense of native 10.5S forms, and their expression of esterase and amidase activity. Fig. 3. Inactivation of esterase and amidase activities of fetal bovine serum AChE exposed to guanidine hydrochloride. Top: Change in esterase and amidase activities with increasing concentrations of guanidine hydrochloride. Bottom: Difference in amidase ⁄ esterase activity ratio. M. F. Montenegro et al. Aryl acylamidase activity of AChE FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 2079 In our attempts to clarify this issue, kinetic con- stants for AChE incubated with 2.2 m guanidine hydrochloride were determined (Table 2). The change in K m and k cat values with ATCh and F-ONA indi- rectly indicated that the catalytic-competent molecules had undergone a conformational rearrangement. Nevertheless, K m values revealed that the change in substrate binding for guanidine hydrochloride-exposed AChE (Table 2) in comparison with untreated enzyme (Table 1) is stronger for ATCh than for F-ONA. The k cat ⁄ K m value for ATCh hydrolysis was reduced 3700- fold by guanidine hydrochloride treatment, whereas that for F-ONA hydrolysis was reduced only 800-fold. This observation supported a more pronounced lower- ing of catalytic efficiency for the former substrate. In addition, the ability of the denaturant guanidine hydrochloride to convert AChE tetramers into dimers [29] made it a useful tool with which to study the rela- tionship between AAA activity and the quaternary organization of AChE. With this aim, samples were incubated with 2.3 m guanidine hydrochloride and later applied to sucrose gradients containing 0.5% polyoxyethylene 10 -oleyl ether (Brij 96). Restoration of native quaternary structure in AChE was prevented by adding 0.5 m guanidine hydrochloride to the gradients. Sedimentation profiles of ATCh-degrading and F-ONA-degrading activities gave a prevalent 9.3S peak with a shoulder at about 10.5S (Fig. 4). This indicated that, although a few guanidine hydrochloride-resistant AChE molecules remained in the native tetrameric state (10.5S), the majority of the tetramers contained partly unfolded subunits (9.3S) [29]. Identification of protein aggregates displaying AChE activity at the tube bottom (Fig. 4) supported exposure of hydropho- bic domains in AChE, a feature exhibited by proteins in a molten globule structure [29,31]. Expression of esterase and amidase activities by AChE tetramers composed of partly unfolded subunits supported the maintenance of both activities in these conditions, in which the native conformation of the polypeptide is severely disturbed. It is probable that tight packing of subunits in tetrameric AChE may restrict oscillations of the polypeptide backbone, thus contributing to a low level of activity under such strong denaturing conditions. Our results are consistent with previous observations suggesting that catalytic efficiency and inhibitor effects are due to global fluctuations of AChE domains [30,32], and varying conformational states of AChE subunits in tetrameric components [33,34]. In our view, a flexible subunit may also be required for maintaining catalytic ability after post-transcriptional linkage of the structural subunits proline-rich membrane anchor (PRiMA) and collagen-like tail subunit (ColQ) to tetramers. In summary, the results reported herein indicate that: (a) the use of F-ONA for measuring the AAA activity of ChEs is advisable; (b) as for BuChE, kinetic parameters show that AChE can degrade the amidase substrates F-ONA and ONA less efficiently than the esterase substrate ATCh; (c) the esterase and amidase activities of AChE tetramers consisting of partly unfolded subunits demonstrate the maintenance of catalytic ability in conditions that change protein structure; and (d) owing to the conformational plasti- city of AChE, the tetramers maintain catalytic com- petence even if the subunit structure differs from the preferred one. Experimental procedures Materials Fetal bovine serum, edrophonium chloride, epoxy-activated agarose, ATCh, 5,5¢-dithiobis(2-nitrobenzoic acid) (Nbs 2 ), tetraisopropyl pyrophosphoramide (Iso-OMPA), BW, Brij 96, guanidine hydrochloride and protein markers for sedi- mentation analysis (beef liver catalase and beef intestine alkaline phosphatase) were all purchased from Sigma (St Louis, MO, USA). o-Nitroaniline and ONA were provided by Merck (Darmstadt, Germany). F-ONA was purchased from Princeton BioMolecular Research (USA), and Fas2 from Latoxan (Valence, France). Echothiophate iodide was donated by Levallois-Perret (France). Pro- tein G-agarose was provided by Boehringer Mannheim (Germany), and the monoclonal antibody HR2 against human brain AChE was provided by Affinity Bioreagents (Golden, CO, USA). Purification of AChE and assay of esterase and amidase activities AChE was affinity-purified from fetal bovine serum using an edrophonium–Sepharose matrix [29]. AChE activity was Table 2. Kinetic parameters of AChE preincubated with 2.2 M gua- nidine hydrochloride. Prior to guanidine hydrochloride incubation, the concentration of active sites was lower in assays with ATCh (approximately 1750 n M) than in those with F-ONA (about 8250 n M). After exposure to 2.2 M guanidine hydrochloride, the concentrations of active sites fell to 87 n M and 412 nM, respec- tively. Substrate K m (mM) k cat (min )1 ) k cat ⁄ K m (M )1 Æmin )1 ) ATCh 1.51 ± 0.12 161 ± 12.8 107 ± 17.1 · 10 3 F-ONA 0.56 ± 0.10 12 ± 2.1 21 ± 7.4 · 10 3 Aryl acylamidase activity of AChE M. F. Montenegro et al. 2080 FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS assayed by the Ellman method as described previously [35], using 1 mm ATCh and 0.33 mm Nbs 2 in 100 mm sodium phosphate (pH 7.5). Purified fetal bovine serum AChE was able to hydrolyze 574 ± 39.5 lmol of ATCh per mg protein per min. Normal esterase and amidase (see below) assays – but not those shown in Fig. 1 – were performed both with and without 10 lm BW, and in the presence of 100 lm Iso- OMPA. The values given correspond to esterase or amidase activities inhibited by BW. In the presence of BW, esterase and amidase activities of purified fetal bovine serum AChE were negligible. Nevertheless, although the samples contained no BuChE, assays included the BuChE inhibitor Iso-OMPA in order to maintain our established assay method, which has been used elsewhere for measuring samples with AChE and BuChE activities. One milliunit (mU) of esterase activity is the amount of enzyme that degrades one nmol of ATCh per min at room temperature (20–25 ° C). AChE activity in fractions col- lected from sucrose gradients was determined by a micro- titer assay [36,37], in which case the activity is also given in milliunits, defined as above but taking into account the volume of sample loaded onto the gradient [12]. Amidase activity was assayed using ONA and F-ONA [12]. In assays with ONA, sample (25–75 l L) and a variable volume (50–0 lL) of 50 mm potassium phosphate (pH 7.0) (to a final volume of 75 lL) were added to microwell plates. The reaction was started by adding 10 mm ONA (150 lL of a 15 mm solution made in the above phosphate buffer and 2% dimethylsulfoxide), and allowed to proceed in an oven at 37 °C. Dimethylsulfoxide was used for its capacity to improve ONA solubility with no effect on ester- ase or amidase activities. The release of o-nitroaniline was recorded at 415 ⁄ 630 nm in a microplate reader (Model 680; Bio-Rad, Hercules, CA, USA) every 15 min for 2–3 h, and amidase activity was calculated by reference to a calibration curve made with 1–200 lm o-nitroaniline. In assays with F-ONA, sample (10–20 lL), 10 lLofa 7.5 mm F-ONA solution made in acetonitrile and enough 50 mm potassium phosphate (pH 7.0) for a total volume of 250 lL were mixed. The concentration of F-ONA in the mixture was 0.3 mm, and the production of o-nitroaniline was recorded at 25 °C. AChE activity was unaffected by the addition of acetonitrile. Amidase activity is given in nmol of ONA or F-ONA hydrolyzed per min (mU). For kinetic measurements, the ATCh concentration ran- ged from 0.02 to 10 mm, that of ONA from 1 to 12 mm, and that of F-ONA from 0.05 to 5 mm. Catalytic parame- ters for ATCh hydrolysis were calculated on the assump- tion of Michaelian behavior of AChE. Thus, K m and V max were calculated by simple weighted nonlinear regression of the Michaelis–Menten equation using the sigma plot pro- gram. Owing to the low solubility of ONA and F-ONA in water, and the poor affinity of AChE for them, the hydro- lysis kinetics were first order in substrate concentration ([S] < K m ). Therefore, kinetic parameters were calculated by fitting the data to the Lineweaver–Burk algorithm with a linear regression program. The concentration of active sites in purified fetal bovine serum AChE was determined with echothiophate iodide as tritrant [26], and the pro- tein content was determined with the Bradford reagent (Bio-Rad). Absence of contaminating esterases from affinity-purified AChE Because of their esterase activity, paraoxonase, carboxyles- terase and serum albumin hydrolyze ONA and F-ONA [38,39], and their presence would therefore lead to overesti- mations of AChE-derived and BuChE-derived amidase activity. Unwanted esterases in samples can be detected by comparing the effects of AChE inhibitors on the hydrolysis of ATCh and acetanilides. Accordingly, the possible pres- ence of contaminating esterases in fetal bovine serum AChE was tested by examining the effects of BW and Fas2 on esterase and amidase activities. Prior to analysis, samples were incubated for 15 min with 1 nm to 10 m m BW (final concentration) or with 2–500 nm Fas2. In addition, the direct relationship between esterase and amidase activities was veri- fied by assaying amidase activity in samples immunodepleted of AChE. With this aim, sample aliquots of fetal bovine serum AChE were incubated with varying amounts of HR2 antibody against AChE bound to protein G–agarose (up to 5 lL of antibody per 50 lL of protein G–agarose). We have previously reported that HR2 recognizes asymmetric, tetra- meric, dimeric and monomeric AChE of human lymph nodes [40] and gut [12,41]. After incubation with HR2, agarose-bound proteins were removed, and samples devoid of AChE were assayed for amidase activity. As an excep- tion, these three types of assay were performed without Iso-OMPA. The use of inhibitors and antibodies allowed us to verify the complete absence of BuChE and unwanted esterases from affinity-purified AChE. Velocity sedimentation analysis Molecular components of AChE were resolved by centri- fugation analysis and identified by their sedimentation coefficients. Samples and sedimentation markers, catalase (11.4S 20,w ; Svedberg units) and alkaline phosphatase (6.1S), were centrifuged at 165 000 g (35 000 r.p.m.) for 20 h at 4°C, on 5-20% sucrose gradients made with 0.5% Brij 96 [35]. Fractions were collected and assayed for sedimentation mark- ers, and AChE and amidase activities [12]. Incubation with guanidine hydrochloride Possible differences in the susceptibility of ATCh-hydro- lyzing and F-ONA-hydrolyzing activities of AChE to the M. F. Montenegro et al. Aryl acylamidase activity of AChE FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 2081 denaturant guanidine hydrochloride were tested by sample incubation with 0.05–3.0 m guanidine hydrochloride. The higher resistance of amidase than of esterase activity to mild guanidine hydrochloride treatment, along with its capacity for dissociating AChE subunits into tetramers [29], led us to study the quaternary structure adopted by the guanidine hydrochloride-resistant AChE molecules. For this, fetal bovine serum AChE was incubated with 2.3 m guanidine hydrochloride and subjected to sedimentation analysis. Sucrose gradients contained 0.5 m guanidine hydrochloride, 0.5% Brij 96, and 5–20% sucrose. Deposi- tion of AChE aggregates at the tube bottom was prevented by placing a 0.5 mL cushion of 40% sucrose on the tube before adding the sucrose gradient. After centrifugation, fractions were collected and assayed for esterase and ami- dase activities, using ATCh and F-ONA as the substrates. Acknowledgements This research was supported by the Fondo de Investiga- cio ´ n Sanitaria of Spain (Grant PI04 ⁄ 1504) and the Fundacio ´ nSe ´ neca of Murcia, Spain (Grant 00636 ⁄ PI ⁄ 04). M. F. Montenegro is a holder of a scholarship from the Fundacio ´ nSe ´ neca. References 1 George ST & Balasubramanian AS (1981) The aryl acylamidase and their relationship to cholinesterase in human serum, erythrocyte and liver. Eur J Biochem 121, 177–186. 2 Balasubramanian AS & Bhanumathy CD (1993) Non- cholinergic functions of cholinesterases. FASEB J 7, 1354–1358. 3 Masson P, Froment MT, Gillon E, Nachon F, Lockridge O & Schopfer LM (2008) Kinetic analysis of effector modulation of butyrylcholinesterase-catalysed hydrolysis of acetanilides and homologous esters. FEBS J 275, 2617–2631. 4 Allebrandt KV, Rajesh V, Andermann P & Layer PG (2005) Expression of acetylcholinesterase (AChE) and aryl acylamidase (AAA) during early zebrafish embryogenesis. Chem Biol Interact 157-158, 353– 355. 5 Boopathy R & Layer PG (2004) Aryl acylamidase activity on acetylcholinesterase is high during early chicken brain development. Protein J 23, 325–333. 6 Wright CI, Geula C & Mesulam MM (1993) Protease inhibitors and indoleamines selectively inhibit cholines- terases in the histopathologic structures of Alzheimer’s disease. Proc Natl Acad Sci USA 90, 683–686. 7 Massoulie ´ J (2002) The origin of the molecular diversity and functional anchoring of cholinesterases. Neuro- signals 11, 130–143. 8 Massoulie ´ J, Perrier N, Noureddine H, Liang D & Bon S (2008) Old and new questions about cholinesterases. Chem Biol Interact 175, 30–44. 9 Meshorer E, Toiber D, Zurel D, Sahly I, Dori A, Cagnano E, Schreiber L, Grisaru D, Tronche F & Soreq H (2004) Combinatorial complexity of 5¢ alternative acetylcholinesterase transcripts and protein products. J Biol Chem 279, 29740–29751. 10 Kryger G, Harel M, Giles K, Toker L, Velan B, Lazar A, Kronman C, Barak D, Ariel N, Shafferman A et al. (2000) Structures of recombinant native and E202Q mutant human acetylcholinesterase complexed with the snake-venom toxin fasciculin-II. Acta Crystallogr D 56, 1385–1394. 11 Nicolet Y, Lockridge O, Masson P, Fontecilla-Camps JC & Nachon F (2003) Crystal structure of human butyrylcholinesterase and of its complexes with substrate and products. J Biol Chem 278, 41141– 41147. 12 Montenegro MF, Moral-Naranjo MT, Pa ´ ez de la Cadena M, Campoy FJ, Mun ˜ oz-Delgado E & Vidal CJ (2008) Human butyrylcholinesterase components differ in aryl acylamidase activity. Biol Chem 389, 425– 432. 13 Weitnauer E, Robitzki A & Layer PG (1998) Aryl acylamidase activity exhibited by butyrylcholinesterase is higher in chick than in horse, but much lower than in fetal calf serum. Neurosci Lett 254, 153–156. 14 Boopathy R, Rajesh RV, Darvesh S & Layer PG (2007) Human serum cholinesterase from liver pathological samples exhibit highly elevated aryl acylamidase activ- ity. Clin Chim Acta 380, 151–156. 15 Moore DH & Hess GP (1975) Acetylcholinesterase- catalyzed hydrolysis of an amide. Biochemistry 14, 2386–2389. 16 Oommen A & Balasubramanian AS (1979) The association of the serotonin-sensitive aryl acylamidase with acetylcholinesterase in the monkey brain. Eur J Biochem 94, 135–143. 17 Darvesh S, McDonald RS, Darvesh KV, Mataija D, Mothana S, Cook H, Carneiro KM, Richard N, Walsh R & Martin E (2006) On the active site for hydrolysis of aryl amides and choline esters by human cholinester- ases. Bioorg Med Chem 14, 4586–4599. 18 Sihotang K (1974) A simple method for purification of acetylcholinesterase from human erythrocyte mem- branes. Biochim Biophys Acta 370, 468–476. 19 Massoulie ´ J & Bon S (1976) Affinity chromatography of acetylcholinesterase. The importance of hydrophobic interations. Eur J Biochem 68, 531–539. 20 Ordentlich A, Barak D, Kronman C, Ariel N, Segall Y, Velan B & Shafferman A (1998) Functional characteris- tics of the oxyanion hole in human acetylcholinesterase. J Biol Chem 273, 19509–19517. Aryl acylamidase activity of AChE M. F. Montenegro et al. 2082 FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 21 Wallace KB & Kemp JR (1991) Species specificity in the chemical mechanisms of organophosphorous anti- cholinesterase activity. Chem Res Toxicol 4, 41–49. 22 Meflah K, Bernard S & Massoulie J (1984) Interactions with lectins indicate differences in the carbohydrate composition of the membrane-bound enzymes acetyl- cholinesterase and 5¢-nucleotidase in different cell types. Biochimie 66, 59–69. 23 Costagli C & Galli A (1998) Inhibition of cholinester- ase-associated aryl acylamidase activity by anticholines- terase agents: focus on drugs potentially effective in Alzheimer’s disease. Biochem Pharmacol 55, 1733–1777. 24 Bourne Y, Taylor P, Radic Z & Marchot P (2003) Structural insights into ligand interactions at the acetyl- cholinesterase peripheral anionic site. EMBO J 22, 1–12. 25 Szegletes T, Mallender WD, Thomas PJ & Rosenberry TL (1999) Substrate binding to the peripheral site of acetylcholinesterase initiates enzymatic catalysis. Sub- strate inhibition arises as a secondary effect. Biochemis- try 38, 122–133. 26 Masson P, Froment MT, Gillon E, Nachon F, Darvesh S & Schopfer LM (2007) Kinetic analysis of butyrylcho- linesterase-catalyzed hydrolysis of acetanilides. Biochim Biophys Acta 1774, 1139–1147. 27 Dura ´ n R, Cerven ˜ ansky C, Dajas F & Tipton KF (1994) Fasciculin inhibition of acetylcholinesterase is prevented by chemical modification of the enzyme at the periph- eral site. Biochim Biophys Acta 1201, 381–388. 28 Gao D & Zhan CG (2005) Modeling effects of oxyan- ion hole on the ester hydrolysis catalyzed by human cholinesterases. J Phys Chem B 109, 23070–23076. 29 Flores-Flores C, Martı ´ nez-Martı ´ nez A, Mun ˜ oz-Delgado E & Vidal CJ (1996) Conversion of acetylcholinesterase hydrophilic tetramers into amphiphilic dimers and monomers. Biochem Biophys Res Commun 219, 53–58. 30 Gorfe AA, Chang CA, Ivanov I & McCammon JA (2008) Dynamics of the acetylcholinesterase tetramer. Biophys J 94, 1144–1154. 31 Shin I, Silman I & Weiner LM (1996) Interaction of partially unfolded forms of Torpedo acetylcholinesterase with liposomes. Protein Sci 5, 42–51. 32 Boyd AE, Dunlop CS, Wong L, Radic Z, Taylor P & Johnson DA (2004) Nanosecond dynamics of acetyl- cholinesterase near the active center gorge. J Biol Chem 279, 26612–26618. 33 Bourne Y, Grassi J, Bougis PE & Marchot P (1999) Conformational flexibility of the acetylcholinesterase tetramer suggested by X-ray crystallography. J Biol Chem 274, 30370–30376. 34 Millard CB, Shnyrov VL, Newstead S, Shin I, Roth E, Silman I & Weiner L (2003) Stabilization of a metasta- ble state of Torpedo californica acetylcholinesterase by chemical chaperones. Protein Sci 12, 2337–2347. 35 Ruiz-Espejo F, Cabezas-Herrera J, Illana J, Campoy FJ & Vidal CJ (2002) Cholinesterase activity and acetyl- cholinesterase glycosylation are altered in human breast cancer. Breast Cancer Res Treat 72, 11–22. 36 Cabezas-Herrera J, Moral-Naranjo MT, Campoy FJ & Vidal CJ (1997) Glycosylation of acetylcholinesterase forms in microsomal membranes from normal and dystrophic Lama2dy mouse muscle. J Neurochem 69, 1964–1974. 37 Moral-Naranjo MT, Campoy FJ, Cabezas-Herrera J & Vidal CJ (1999) Increased butyrylcholinesterase levels in microsomal membranes of dystrophic Lama2dy mouse muscle. J Neurochem 73, 1138–1144. 38 Li B, Sedlacek M, Manoharan I, Boopathy R, Duysen EG, Masson P & Lockridge O (2005) Butyrylcholine- sterase, paraoxonase, and albumin esterase, but not carboxylesterase, are present in human plasma. Biochem Pharmacol 70, 1673–1684. 39 Masson P, Froment MT, Darvesh S, Schopfer LM & Lockridge O (2007) Aryl acylamidase activity of human serum albumin with o-nitrotrifluoroacetanilide as the substrate. J Enzyme Inhib Med Chem 22, 463–469. 40 Ruiz-Espejo F, Cabezas-Herrera J, Illana J, Campoy FJ, Mun ˜ oz-Delgado E & Vidal CJ (2003) Breast cancer metastasis alters acetylcholinesterase activity and the composition of enzyme forms in axillary lymph nodes. Breast Cancer Res Treat 80, 105–114. 41 Montenegro MF, Ruiz-Espejo F, Campoy FJ, Mun ˜ oz- Delgado E, Pa ´ ez de la Cadena M, Rodrı ´ guez-Berrocal FJ & Vidal CJ (2006) Cholinesterases are down- expressed in human colorectal carcinoma. Cell Mol Life Sci 63, 2175–2182. 42 Masson P, Legrand P, Bartels CF, Froment MT, Schopfer LM & Lockridge O (1997) Role of aspar- tate 70 and tryptophan 82 in binding of succinyldithi- ocholine to human butyrylcholinesterase. Biochemistry 36, 2266–2277. M. F. Montenegro et al. Aryl acylamidase activity of AChE FEBS Journal 276 (2009) 2074–2083 ª 2009 The Authors Journal compilation ª 2009 FEBS 2083 . Hydrolysis of acetylthiocoline, o-nitroacetanilide and o-nitrotrifluoroacetanilide by fetal bovine serum acetylcholinesterase Marı ´ a. BuChE species (G H 4 ) of human plasma and AChE forms of fetal bovine serum prompted us to study the esterase and amidase activities of fetal bovine serum AChE. The